Self-learning - Spatial Transcriptomics

Learning Objectives

  • Experimental design best practices of current spatial transcriptomics technologies
  • Data loading and quality control of Visium HD samples
  • Standard single-cell workflows adapted for spatial data, including highly variable gene selection, PCA, UMAP, kNN, and clustering
  • Cell type annotation with deconvolution using RCTD for sequencing-based technologies
  • Spatial-specific analyses that utilize the physical location of bins/cells on the tissue, including:
    • Spatial clustering with BANKSY
    • Spatially variable gene detection with Moran’s I
    • Cell-cell communication with CellChat

Installations

Follow the installation instructions on the main page.

Lessons

Part 1: Pre-reading

  1. Spatial Technology Overview
  2. Space Ranger Summary

Part 2: Set-up and QC

  1. Load Visium HD Data
  2. Theory of PCA
  3. Quality Control

Part 3: scRNA-seq workflow

  1. Normalization and Sketch Downsampling
  2. Dimensionality Reduction
  3. Clustering
  4. Integration
  5. Seurat Cheatsheet

Part 4: Spatial and coordinate-based analyses

  1. BANKSY Spatial Clustering
  2. Deconvolution
  3. Differential Expression and Pathway Analysis
  4. Spatially Variable Genes
  5. Cell-Cell Communication

Resources